{"id":"5a3a68b6-f250-4eb0-a2f2-9a100fd6801c","shortId":"cgS4C9","kind":"skill","title":"brain-jam-plan","tagline":"Use when running claudikins-kernel:outline, brainstorming implementation approaches, gathering requirements iteratively, structuring complex technical plans, or facing analysis paralysis with too many options — provides iterative human-in-the-loop planning with explicit checkpoin","description":"# Brain-Jam Planning Methodology\n\n## Overview\n\nPlanning is an iterative conversation, not a production line. The human stays in the loop at every phase.\n\n> \"Go back and forth with Claude until I like its plan. A good plan is really important.\" — Boris\n\n**Core principle:** Understand before solving. Ask before assuming. Recommend but let user decide.\n\n## Trigger Symptoms\n\nUse this skill when:\n\n- Running the `claudikins-kernel:outline` command\n- Requirements are unclear or keep changing\n- Multiple valid approaches exist and you can't choose\n- User wants involvement in decisions (not just receive a plan)\n- Previous plans failed due to missed requirements\n- Scope creep is a concern\n- You're tempted to just start coding without a plan\n\n## The Brain-Jam Process\n\n### Phase 1: Requirements Gathering\n\n**One question at a time. Wait for the answer.**\n\nKey questions to answer:\n\n1. What problem are we solving?\n2. What does success look like?\n3. What constraints exist?\n4. What's explicitly OUT of scope?\n\nUse `AskUserQuestion` with specific options — never open-ended unless necessary.\n\n### Phase 2: Context Building\n\nBefore proposing solutions, understand the landscape:\n\n- What exists already in the codebase?\n- What patterns should we follow?\n- What dependencies apply?\n- What has been tried before?\n\n### Phase 3: Approach Generation\n\nGenerate 2-3 distinct approaches. Each must include:\n\n| Element | Purpose               |\n| ------- | --------------------- |\n| Summary | 1-2 sentence overview |\n| Pros    | Clear benefits        |\n| Cons    | Honest trade-offs     |\n| Effort  | low / medium / high   |\n| Risk    | low / medium / high   |\n\n**Always recommend one with reasoning.** See [approach-template.md](references/approach-template.md).\n\n### Phase 4: Section-by-Section Drafting\n\nDraft one section at a time. Get approval before moving on.\n\n**Never batch approvals** — each checkpoint is a chance to course-correct.\n\n## Non-Negotiables\n\nThese rules have no exceptions:\n\n**One question at a time.**\n\n- Not \"let me ask a few things\"\n- Not \"quick questions\"\n- One question. Wait. Process answer. Next question.\n\n**Always present 2-3 approaches.**\n\n- Not \"here's what I recommend\"\n- Not \"the obvious choice is...\"\n- Present options. Recommend one. User decides.\n\n**Checkpoint before proceeding.**\n\n- Not \"I'll assume that's fine\"\n- Not \"continuing unless you object\"\n- Explicit approval. \"Does this look right?\" Wait for yes.\n\n**Never fabricate research.**\n\n- Not \"based on my understanding\"\n- Not \"typically in codebases like this\"\n- If you don't know, research or ask. Don't invent.\n\n## Plan Quality Criteria\n\nA good plan has:\n\n- [ ] Clear problem statement\n- [ ] Explicit scope boundaries (IN and OUT)\n- [ ] Measurable success criteria\n- [ ] Task breakdown with dependencies\n- [ ] Risk identification and mitigations\n- [ ] Verification checklist\n\nSee [plan-checklist.md](references/plan-checklist.md) for full verification.\n\n## Output Format\n\nPlans must include machine-readable task markers for `claudikins-kernel:execute` compatibility:\n\n```markdown\n<!-- EXECUTION_TASKS_START -->\n\n| #   | Task          | Files                | Deps | Batch |\n| --- | ------------- | -------------------- | ---- | ----- |\n| 1   | Create schema | prisma/schema.prisma | -    | 1     |\n| 2   | Add service   | src/services/user.ts | 1    | 1     |\n\n<!-- EXECUTION_TASKS_END -->\n```\n\nSee [plan-format.md](references/plan-format.md) for complete structure.\n\n## Anti-Patterns\n\n**Don't do these:**\n\n- Batching multiple questions together\n- Proposing solutions before understanding requirements\n- Presenting only one approach\n- Skipping the verification checklist\n- Continuing without explicit approval at checkpoints\n- Fabricating research findings when data is sparse\n\n## Rationalizations to Resist\n\nAgents under pressure find excuses. These are all violations:\n\n| Excuse                                                | Reality                                                       |\n| ----------------------------------------------------- | ------------------------------------------------------------- |\n| \"I'll batch questions to save time\"                   | Batching causes missed requirements. One at a time.           |\n| \"User knows what they want, skip brain-jam\"           | Assumptions cause rework. Gather requirements explicitly.     |\n| \"I'll propose solutions while gathering requirements\" | Solutions bias requirements. Understand first, solve second.  |\n| \"User implied preference, don't need alternatives\"    | Implied ≠ decided. Always present 2-3 options.                |\n| \"This is simple, don't need checkpoints\"              | Simple plans still fail. Checkpoints catch errors early.      |\n| \"I already know the right approach\"                   | Your confidence isn't approval. User decides.                 |\n| \"Alternatives will confuse them\"                      | Confusion means requirements are unclear. Clarify.            |\n| \"I'll get approval for multiple sections at once\"     | Batched approvals hide problems. One section, one checkpoint. |\n\n**All of these mean: Follow the methodology. No shortcuts.**\n\n## Red Flags — STOP and Reassess\n\nIf you're thinking any of these, you're about to violate the methodology:\n\n- \"Let me just quickly...\"\n- \"The user probably wants...\"\n- \"This is obvious, I don't need to ask\"\n- \"I'll come back to requirements later\"\n- \"One approach is clearly best\"\n- \"They already approved something similar\"\n- \"Checkpoints slow things down\"\n- \"I know what they meant\"\n\n**All of these mean: STOP. Return to methodology. Ask, don't assume.**\n\n## Edge Case Handling\n\n| Situation                        | Reference                                                                     |\n| -------------------------------- | ----------------------------------------------------------------------------- |\n| Context collapse mid-plan        | [session-collapse-recovery.md](references/session-collapse-recovery.md)       |\n| Endless iteration loop           | [iteration-limits.md](references/iteration-limits.md)                         |\n| Research taking too long         | [research-timeouts.md](references/research-timeouts.md)                       |\n| Approaches contradict each other | [approach-conflict-resolution.md](references/approach-conflict-resolution.md) |\n| User abandons plan               | [plan-abandonment-cleanup.md](references/plan-abandonment-cleanup.md)         |\n| Requirements keep changing       | [requirement-stability.md](references/requirement-stability.md)               |\n\n## References\n\n- [plan-checklist.md](references/plan-checklist.md) — Verification checklist\n- [approach-template.md](references/approach-template.md) — How to present options\n- [plan-format.md](references/plan-format.md) — Output structure for claudikins-kernel:execute\n- [iteration-limits.md](references/iteration-limits.md) — When to stop iterating\n- [requirement-stability.md](references/requirement-stability.md) — Scope creep detection\n- [approach-conflict-resolution.md](references/approach-conflict-resolution.md) — Conflicting approaches\n- [session-collapse-recovery.md](references/session-collapse-recovery.md) — Context collapse handling\n- [research-timeouts.md](references/research-timeouts.md) — Timeout handling\n- [plan-abandonment-cleanup.md](references/plan-abandonment-cleanup.md) — Cleanup procedures","tags":["brain","jam","plan","claudikins","kernel","elb-pr","agent-skills","claude","claude-ai","claude-code","claude-code-plugin","claude-skills"],"capabilities":["skill","source-elb-pr","skill-brain-jam-plan","topic-agent-skills","topic-claude","topic-claude-ai","topic-claude-code","topic-claude-code-plugin","topic-claude-skills","topic-claudecode","topic-sre"],"categories":["claudikins-kernel"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/elb-pr/claudikins-kernel/brain-jam-plan","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add elb-pr/claudikins-kernel","source_repo":"https://github.com/elb-pr/claudikins-kernel","install_from":"skills.sh"}},"qualityScore":"0.508","qualityRationale":"deterministic score 0.51 from registry signals: · indexed on github topic:agent-skills · 117 github stars · SKILL.md body (7,374 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-04-22T06:55:19.083Z","embedding":null,"createdAt":"2026-04-20T07:27:59.247Z","updatedAt":"2026-04-22T06:55:19.083Z","lastSeenAt":"2026-04-22T06:55:19.083Z","tsv":"'-2':257 '-3':247,347,595 '1':162,178,256,471,475,480,481 '2':184,213,246,346,476,594 '3':190,242 '4':194,285 'abandon':765 'add':477 'agent':528 'alreadi':224,613,710 'altern':589,625 'alway':276,344,592 'analysi':24 'answer':173,177,341 'anti':489 'anti-pattern':488 'appli':235 'approach':14,117,243,249,348,507,617,705,758,808 'approach-conflict-resolution.md':762,805 'approach-template.md':282,779 'approv':298,304,382,515,622,638,645,711 'ask':88,330,411,696,731 'askuserquest':202 'assum':90,372,734 'assumpt':563 'back':66,700 'base':394 'batch':303,470,495,541,546,644 'benefit':262 'best':708 'bias':577 'bori':82 'boundari':427 'brain':2,42,158,561 'brain-jam':41,157,560 'brain-jam-plan':1 'brainstorm':12 'breakdown':435 'build':215 'case':736 'catch':609 'caus':547,564 'chanc':309 'chang':114,771 'checklist':443,511,778 'checkpoin':40 'checkpoint':306,366,517,603,608,651,714 'choic':358 'choos':123 'clarifi':634 'claud':70 'claudikin':9,105,462,791 'claudikins-kernel':8,104,461,790 'cleanup':820 'clear':261,422,707 'code':152 'codebas':227,401 'collaps':741,812 'come':699 'command':108 'compat':465 'complet':486 'complex':19 'con':263 'concern':145 'confid':619 'conflict':807 'confus':627,629 'constraint':192 'context':214,740,811 'continu':377,512 'contradict':759 'convers':51 'core':83 'correct':313 'cours':312 'course-correct':311 'creat':472 'creep':142,803 'criteria':417,433 'data':522 'decid':95,365,591,624 'decis':128 'dep':469 'depend':234,437 'detect':804 'distinct':248 'draft':290,291 'due':137 'earli':611 'edg':735 'effort':268 'element':253 'end':209 'endless':747 'error':610 'everi':63 'except':321 'excus':532,537 'execut':464,793 'exist':118,193,223 'explicit':39,197,381,425,514,568 'fabric':391,518 'face':23 'fail':136,607 'file':468 'find':520,531 'fine':375 'first':580 'flag':662 'follow':232,656 'format':451 'forth':68 'full':448 'gather':15,164,566,574 'generat':244,245 'get':297,637 'go':65 'good':77,419 'handl':737,813,817 'hide':646 'high':271,275 'honest':264 'human':33,57 'human-in-the-loop':32 'identif':439 'implement':13 'impli':584,590 'import':81 'includ':252,454 'invent':414 'involv':126 'isn':620 'iter':17,31,50,748,799 'iteration-limits.md':750,794 'jam':3,43,159,562 'keep':113,770 'kernel':10,106,463,792 'key':174 'know':408,555,614,719 'landscap':221 'later':703 'let':93,328,680 'like':73,189,402 'line':55 'll':371,540,570,636,698 'long':755 'look':188,385 'loop':36,61,749 'low':269,273 'machin':456 'machine-read':455 'mani':28 'markdown':466 'marker':459 'mean':630,655,726 'meant':722 'measur':431 'medium':270,274 'methodolog':45,658,679,730 'mid':743 'mid-plan':742 'miss':139,548 'mitig':441 'move':300 'multipl':115,496,640 'must':251,453 'necessari':211 'need':588,602,694 'negoti':316 'never':206,302,390 'next':342 'non':315 'non-negoti':314 'object':380 'obvious':357,690 'off':267 'one':165,278,292,322,337,363,506,550,648,650,704 'open':208 'open-end':207 'option':29,205,361,596,784 'outlin':11,107 'output':450,787 'overview':46,259 'paralysi':25 'pattern':229,490 'phase':64,161,212,241,284 'plan':4,21,37,44,47,75,78,133,135,155,415,420,452,605,744,766 'plan-abandonment-cleanup.md':767,818 'plan-checklist.md':445,775 'plan-format.md':483,785 'prefer':585 'present':345,360,504,593,783 'pressur':530 'previous':134 'principl':84 'prisma/schema.prisma':474 'probabl':686 'problem':180,423,647 'procedur':821 'proceed':368 'process':160,340 'product':54 'propos':217,499,571 'pros':260 'provid':30 'purpos':254 'qualiti':416 'question':166,175,323,336,338,343,497,542 'quick':335,683 'ration':525 're':147,668,674 'readabl':457 'realiti':538 'realli':80 'reason':280 'reassess':665 'receiv':131 'recommend':91,277,354,362 'red':661 'refer':739,774 'references/approach-conflict-resolution.md':763,806 'references/approach-template.md':283,780 'references/iteration-limits.md':751,795 'references/plan-abandonment-cleanup.md':768,819 'references/plan-checklist.md':446,776 'references/plan-format.md':484,786 'references/requirement-stability.md':773,801 'references/research-timeouts.md':757,815 'references/session-collapse-recovery.md':746,810 'requir':16,109,140,163,503,549,567,575,578,631,702,769 'requirement-stability.md':772,800 'research':392,409,519,752 'research-timeouts.md':756,814 'resist':527 'return':728 'rework':565 'right':386,616 'risk':272,438 'rule':318 'run':7,102 'save':544 'schema':473 'scope':141,200,426,802 'second':582 'section':287,289,293,641,649 'section-by-sect':286 'see':281,444,482 'sentenc':258 'servic':478 'session-collapse-recovery.md':745,809 'shortcut':660 'similar':713 'simpl':599,604 'situat':738 'skill':100 'skill-brain-jam-plan' 'skip':508,559 'slow':715 'solut':218,500,572,576 'solv':87,183,581 'someth':712 'source-elb-pr' 'spars':524 'specif':204 'src/services/user.ts':479 'start':151 'statement':424 'stay':58 'still':606 'stop':663,727,798 'structur':18,487,788 'success':187,432 'summari':255 'symptom':97 'take':753 'task':434,458,467 'technic':20 'tempt':148 'thing':333,716 'think':669 'time':169,296,326,545,553 'timeout':816 'togeth':498 'topic-agent-skills' 'topic-claude' 'topic-claude-ai' 'topic-claude-code' 'topic-claude-code-plugin' 'topic-claude-skills' 'topic-claudecode' 'topic-sre' 'trade':266 'trade-off':265 'tri':239 'trigger':96 'typic':399 'unclear':111,633 'understand':85,219,397,502,579 'unless':210,378 'use':5,98,201 'user':94,124,364,554,583,623,685,764 'valid':116 'verif':442,449,510,777 'violat':536,677 'wait':170,339,387 'want':125,558,687 'without':153,513 'yes':389","prices":[{"id":"46e83bb1-6a98-4950-8d13-113de942789e","listingId":"5a3a68b6-f250-4eb0-a2f2-9a100fd6801c","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"elb-pr","category":"claudikins-kernel","install_from":"skills.sh"},"createdAt":"2026-04-20T07:27:59.247Z"}],"sources":[{"listingId":"5a3a68b6-f250-4eb0-a2f2-9a100fd6801c","source":"github","sourceId":"elb-pr/claudikins-kernel/brain-jam-plan","sourceUrl":"https://github.com/elb-pr/claudikins-kernel/tree/main/skills/brain-jam-plan","isPrimary":false,"firstSeenAt":"2026-04-20T07:27:59.247Z","lastSeenAt":"2026-04-22T06:55:19.083Z"}],"details":{"listingId":"5a3a68b6-f250-4eb0-a2f2-9a100fd6801c","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"elb-pr","slug":"brain-jam-plan","github":{"repo":"elb-pr/claudikins-kernel","stars":117,"topics":["agent-skills","claude","claude-ai","claude-code","claude-code-plugin","claude-skills","claudecode","sre"],"license":"mit","html_url":"https://github.com/elb-pr/claudikins-kernel","pushed_at":"2026-04-21T15:23:37Z","description":"SRE thinking applied to Claude Code, based on Boris Cherny's Q&A. It enforces a strict 4-stage pipeline with gates between each step. You literally cannot skip verification. You cannot ship without approval.","skill_md_sha":"5901852672737922ceaead000528e91177166f24","skill_md_path":"skills/brain-jam-plan/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/elb-pr/claudikins-kernel/tree/main/skills/brain-jam-plan"},"layout":"multi","source":"github","category":"claudikins-kernel","frontmatter":{"name":"brain-jam-plan","description":"Use when running claudikins-kernel:outline, brainstorming implementation approaches, gathering requirements iteratively, structuring complex technical plans, or facing analysis paralysis with too many options — provides iterative human-in-the-loop planning with explicit checkpoints and trade-off presentation"},"skills_sh_url":"https://skills.sh/elb-pr/claudikins-kernel/brain-jam-plan"},"updatedAt":"2026-04-22T06:55:19.083Z"}}